Nicholas Albion
Nicholas Albion
I've done this in PR #65
I've changed my window size to 8ms and removed the "+1" mentioned above, but now when FFT returns the first element always a 0.0 imaginary component, and as a result...
@Skylion007 @lkuza2 @ClusterM I'm working on this here - https://github.com/nalbion/java-speech-api/blob/feature/lex-recogniser/src/main/java/com/darkprograms/speech/recognizer/vad/VoiceActivityDetector.java I'm trying to build a continuously listening service with recognition requests triggered by a VAD. Any tips/assistance would be welcome
@Skylion007 sorry, no. I think that there's something wrong with my microphone(s). I don't think that the algorithms in the VAD are correct either.
@brmc did you figure out how to get it to build? I have the same problem in #51 and have reproduced the problem in Docker. It would be good to...
I've made a start on this: https://github.com/OpenASR/idear/commit/a945ad405eb7e5a69d5f3abf0c306966d93372d2
Actually @kenarsa, I am a bit concerned about the limitations of Picovoice: > 5 contexts, each with a limit of 50 unique words and 150 total words, 5 intents, and...
I'm using Windows x64
@breandan how would you provide the `.so`/`.dll` libraries with the plugin?
There is a pending PR which will resolve this - dodo/node-unicodetable#22